Cost Shifting is Not Cost Saving

August 24th, 2011 at 6:52 pm

There are a few very important themes that we should all keep in our heads when we’re debating changes to America’s system of health care provision.

First, the economic pressure from health care spending is not a government problem, it’s a systemic problem.  If anything, spending is growing faster in the private than the public sector.  Here’s a graph I posted a while ago that is highly germane to this post too, as you’ll see in a moment.

Second, and this is most important in today’s climate of cut, cut, cut: cost shifting is not cost saving.

When Rep Ryan argued he could save Medicare by providing a voucher that was worth a lot less than the cost of comparable coverage, he was shifting, not saving, on costs.  And that’s also what we would be doing if we increased the eligibility age of Medicare from 65 to 67, an option that may well be put before the Congressional super-committee tasked with more deficit reduction.

My CBPP colleague Paul Van de Water has a new paper out today on this proposal and the analysis features a graph that usefully combines both of the above points.

Medicare is actually a less expensive way to provide coverage to older persons relative to the alternatives.  As Paul notes:

“People who lost Medicare would have to seek health coverage from other sources.  This would affect not only their own personal budgets but also employers’ costs, state budgets, and the premiums paid by Medicare beneficiaries and participants in the new health insurance exchanges.”

As the figure shows, the increase in these other sources of health care would be a lot more expensive than the savings.  Here’s a case where the outcome is worse than principle #2 above: not only do cost shifting not generate cost savings.  It generates higher costs.

Other key findings from Paul’s study:

 65- and 66-year-olds losing Medicare coverage would face higher out-of-pocket health care costs, on average.  Two-thirds of this group — 3.3 million people — would face an average of $2,200 more each year in premiums and cost-sharing charges.

 Employers that provide health coverage to their retirees would face higher costs as more 65- and 66-year-olds received primary coverage through their employer rather than Medicare. Medicare beneficiaries, as well as people under age 65 who buy insurance through the new health insurance exchanges, would face higher premiums as 65- and 66-year-olds left Medicare and many of them bought coverage through the exchanges.

 State Medicaid costs would rise as some of the people who lost Medicare coverage would shift to Medicaid.

So here’s the follow-up study I’d really like to see: how much would we save, system-wide (i.e., private and public combined), if instead of this bad idea of raising Medicare’s eligibility age by two years, we lowered it by two years.

I realize this violates principle #2—my hypothesis in this case is that lowering the Medicare eligibility age would generate savings in that gov’t spending would rise but that increase would be more than offset by the decline in private spending.  But it’s an exception I’m happy to make.

C’mon—there must be some wonk out there with some time on her hands.  And, assuming sound methodology, I will feature your work in an OTE post, even if I’m wrong!


Print Friendly, PDF & Email

11 comments in reply to "Cost Shifting is Not Cost Saving"

  1. Another Matt says:

    Thanks for this.

    I think the argument should be made as simply as possible for it to have any effect in the world:

    If we raise the age of eligibility for medicare, the people who move from medicare to private insurance (and they’ll have to if the individual mandate holds up) will cause the average age of the private risk pool to increase, so due to the dynamics of risk pooling, premiums will be higher. Meanwhile, by pulling younger people out of medicare, medicare’s average age will be increased as well, raising their cost of insurance. The cost per insured person increases across the board. So the logic goes, we could lower the age of eligibility and reduce the average age of both risk pools and thus reduce the cost of insuring everyone. I don’t see a “sweet spot” either — the lower you go, the more savings, in terms of premiums (other factors might come into play).

    It’s also important to point out that this has nothing to do with usual “liberal” or “conservative” political values — it’s only due to the binary nature of our hybrid system. Consider an alternative system where the government insures everyone from ages 0-30, and then when you turn 30 you “graduate” into the private pool. In this scenario, the “conservative” goal of lowering the “graduation age” to 28 to get more people into the private pool and cut government spending actually would save money across the board for the same reasons outlined above — the average age of both pools decreases.

    • Tom says:

      Let’s see if I understand the argument. Raising the eligibility age would increase costs because the average age would increase and implicitly the older one is the more resources one uses. On the contrary reducing the eligiblity age would save money because the each pool would have a lower cost basis. This it seems obvious: Eligibility for medicare should be reduced and the raising health care costs are resolved!

      • Another Matt says:

        That’s right — implicitly the argument should follow from the positive correlation between age and healthcare resource use. One would need to research this — young children may use more healthcare resources than those in their teens and 20s, but typically those children are on their parents’ plan or medicaid anyway.

        Also, it may be a stretch to say that the rising healthcare costs are resolved by such a proposal, since there are many other factors at play; but I do think this contributes.

        • Stanley Seiden says:

          Well, hold on, though, because taking this argument to its conclusion would suggest we should drop the Medicare age over and over again, perpetually lowering average ages ad infinitum. And the risk of lowering the eligible age is that it gives incentive for able bodies to drop out of the work force sooner; it would seem that an ideal system would somehow incentivize staying in the work for for two to four years longer despite enrolling in a Medicare plan at age 63, although I can see no reason why anyone would feel compelled to accept such an offer…

          • Another Matt says:

            “And the risk of lowering the eligible age is that it gives incentive for able bodies to drop out of the work force sooner…”

            This is an extremely good point, but I think this is one of those things that is mostly sensitive to where the boundary is. There are ways of looking at the age of eligibility for medicare as being deliberately tied to retirement, along with social security and the age at which 401(k) money can be withdrawn without penalty. But there are other ways of looking at it that might suggest it’s “merely an accident” that currently has psychological import.

            So consider a system where you could get into medicare at age 50 (say), but still had to wait until 65 for the other retirement stuff to kick in. I think the phenomenon you are talking about would be less of an issue; at the margins, yes, some people will be able to retire earlier because they won’t have to work to pay for their healthcare, but this isn’t exactly a problem in Canada. Me, I would just extend the age of eligibility all the way down to 0, and allow people to pay premiums (“public option”) or just go straight to single-payer.

            Also, I think it’s only a big problem in times of low unemployment, where there might be demand for labor that isn’t being met because people are dropping out of the workforce. But there are worse problems to have than opportunity costs during low unemployment. In times of high unemployment, if a 63-year old drops out of the workforce comfortably and someone 30-50 years old who was previously unemployed replaces them, this is not a bad thing — our entitlement system is structurally dependent on population growth and the productivity of the young. Averaged across the population, I think young idle labor resources is more of a problem than senior idle labor resources.

    • Patricia Carey says:

      Thanks for the very lucid explanation. It’s like a cool drink of water in the desert.

    • Patrick says:

      While I completely agree with Mr. Bernstein’s argument, I think your argument is a logical fallacy. It is similar to Simpson’s paradox.

      It is true that lowering the eligibility age would lower the average age of both pools, and thus lower the average cost of insuring people in each pool. However, the two pools have different average costs, and you will be adding more people to the pool with the higher average cost. Therefore, the total cost in the system does not necessarily decrease. If private and public insurance costs were the same at any given age, then changing the eligibility age would have no effect. If the public insurance costs were higher at a given age, then lower the eligibility would actually increase the overall cost.

      In summary, the savings only arise because public health insurance is better at controlling costs (is more efficient) than private health insurance in the US.

  2. Andrea says:

    Testing, testing, 1, 2, 3.

  3. Michael says:

    Since Republicans consider the vast majority of their fellow Americans a burden to be occasionally looted and always starved, cost shifting which moves expenses from the public arena to individual poor and middle class people — especially women and persons of color — is saving, by definition.

    The basic tenet of conservatism is that we are not all in this together. There are “real Americans” and lesser beings.

    More pithily: the only part of the Constitution which Teabaggers understand and support wholeheartedly is the 3/5ths Compromise.

  4. Geoff Freedman says:

    What the Republicans are simply trying to is to push Medicare into the private sector in order relieve the government of the cost burden and ther unfundedliabilties down the road without dealing with the large unfunded future liability as well as the out of control costs associated with our fee for services based health system. They don’t seem to give a hang about actual health costs themselves,

  5. Another Matt says:

    Sure, I appreciate it! Note that this argument only covers the effects on premium; this paper and JB’s proposed study take other things into account like what happens to the specific 65-66-year-olds who are affected, and those who aren’t eligible for medicare but are for medicaid, and so forth. I also assume that the premiums argument here is a little too cute to describe the real world without some modification at the margins, but I think the logic is basically right.